Import Libraries¶
In [662]:
import time
import os
# Simple file path extractor
import glob
# Matrice Operations
import numpy as np
# Image Processing
import cv2
# Visualizing Images
import matplotlib.pyplot as plt
# Visualizing 3D Plots
from mpl_toolkits.mplot3d import Axes3D
# Display plots in notebook
%matplotlib inline
# Hog Features Extractor
from skimage.feature import hog
# Data Normalization
from sklearn.preprocessing import StandardScaler
# Data Split
from sklearn.cross_validation import train_test_split
# Linear SVM
from sklearn.svm import LinearSVC
from scipy.ndimage.measurements import label
# Will stick with cv2.imread() for scaling consistency
# import matplotlib.image as mpimg
Tools Investigation¶
Experimenting with each of the components that can be used in the vehicle detection pipeline.
Reading Images¶
In [344]:
def imread(image_path):
image = cv2.imread(image_path)
image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)
return image
Drawing Boxes¶
Prepare a method to draw boxes around detected cars. Define a function that takes an image, a list of bounding boxes, and optional color tuple and line thickness as inputs then draws boxes in that color on the output:
In [486]:
def draw_boxes(img, bboxes, color=(0, 0, 255), thick=6):
# Make a copy of the image
imcopy = np.copy(img)
# Iterate through the bounding boxes
for bbox in bboxes:
# Draw a rectangle given bbox coordinates
cv2.rectangle(imcopy, bbox[0], bbox[1], color, thick)
# Return the image copy with boxes drawn
return imcopy
Test boxes drawing
In [487]:
image = imread('test_images/test4.jpg')
# Add bounding boxes in this format, these are just example coordinates.
bboxes = [((800, 400), (950, 500)), ((1030, 400), (1250, 500))]
result = draw_boxes(image, bboxes)
plt.imshow(result)
plt.imsave("output_images/drawing_boxes.png", result)
Histograms of Color¶
The following function computes histogram for each colour channel of an image:
In [488]:
# Define a function to compute color histogram features
def color_hist(img, nbins=32, bins_range=(0, 256)):
# Compute the histogram of the RGB channels separately
rhist = np.histogram(img[:,:,0], bins=nbins, range=bins_range)
ghist = np.histogram(img[:,:,1], bins=nbins, range=bins_range)
bhist = np.histogram(img[:,:,2], bins=nbins, range=bins_range)
# Concatenate the histograms into a single feature vector
hist_features = np.concatenate((rhist[0], ghist[0], bhist[0]))
# If you want to return just the hist_features
# return hist_features
# Generating bin centers
bin_edges = rhist[1]
bin_centers = (bin_edges[1:] + bin_edges[0:len(bin_edges) - 1]) / 2
# Return the individual histograms, bin_centers and feature vector
return rhist, ghist, bhist, bin_centers, hist_features
Test color histograms:
In [489]:
image = imread('test_images/cutout1.jpg')
rh, gh, bh, bincen, feature_vec = color_hist(image, nbins=32, bins_range=(0, 256))
# Plot a figure with all three bar charts
if rh is not None:
fig = plt.figure(figsize=(12,3))
plt.subplot(131)
plt.bar(bincen, rh[0])
plt.xlim(0, 256)
plt.title('R Histogram')
plt.subplot(132)
plt.bar(bincen, gh[0])
plt.xlim(0, 256)
plt.title('G Histogram')
plt.subplot(133)
plt.bar(bincen, bh[0])
plt.xlim(0, 256)
plt.title('B Histogram')
fig.tight_layout()
else:
print('Your function is returning None for at least one variable...')
Exploring Colour Spaces¶
The following function can be used to explore the distribution of colour channels in an image:
In [490]:
def plot3d(pixels, colors_rgb,
axis_labels=list("RGB"), axis_limits=((0, 255), (0, 255), (0, 255))):
"""Plot pixels in 3D."""
# Create figure and 3D axes
fig = plt.figure(figsize=(8, 8))
ax = Axes3D(fig)
# Set axis limits
ax.set_xlim(*axis_limits[0])
ax.set_ylim(*axis_limits[1])
ax.set_zlim(*axis_limits[2])
# Set axis labels and sizes
ax.tick_params(axis='both', which='major', labelsize=14, pad=8)
ax.set_xlabel(axis_labels[0], fontsize=16, labelpad=16)
ax.set_ylabel(axis_labels[1], fontsize=16, labelpad=16)
ax.set_zlabel(axis_labels[2], fontsize=16, labelpad=16)
# Plot pixel values with colors given in colors_rgb
ax.scatter(
pixels[:, :, 0].ravel(),
pixels[:, :, 1].ravel(),
pixels[:, :, 2].ravel(),
c=colors_rgb.reshape((-1, 3)), edgecolors='none')
return ax # return Axes3D object for further manipulation
Check distribution:
In [491]:
# Read a color image
img = imread('test_images/test4.jpg')
# Select a small fraction of pixels to plot by subsampling it
scale = max(img.shape[0], img.shape[1], 64) / 64 # at most 64 rows and columns
img_small = cv2.resize(img, (np.int(img.shape[1] / scale), np.int(img.shape[0] / scale)), interpolation=cv2.INTER_NEAREST)
# Convert subsampled image to desired color space(s)
img_small_RGB = cv2.cvtColor(img_small, cv2.COLOR_BGR2RGB) # OpenCV uses BGR, matplotlib likes RGB
img_small_HSV = cv2.cvtColor(img_small, cv2.COLOR_BGR2HSV)
img_small_rgb = img_small_RGB / 255. # scaled to [0, 1], only for plotting
# Plot and show
plot3d(img_small_RGB, img_small_rgb)
plt.show()
plot3d(img_small_HSV, img_small_rgb, axis_labels=list("HSV"))
plt.show()
Spatial Binning of Color¶
Histograms of raw pixel values can be used as features for image recognition:
In [492]:
# Function to compute color histogram features
# Pass the color_space flag as 3-letter all caps string
# like 'HSV' or 'LUV' etc.
def bin_spatial(img, color_space='RGB', size=(32, 32)):
# Convert image to new color space (if specified)
if color_space != 'RGB':
if color_space == 'HSV':
feature_image = cv2.cvtColor(img, cv2.COLOR_RGB2HSV)
elif color_space == 'LUV':
feature_image = cv2.cvtColor(img, cv2.COLOR_RGB2LUV)
elif color_space == 'HLS':
feature_image = cv2.cvtColor(img, cv2.COLOR_RGB2HLS)
elif color_space == 'YUV':
feature_image = cv2.cvtColor(img, cv2.COLOR_RGB2YUV)
elif color_space == 'YCrCb':
feature_image = cv2.cvtColor(img, cv2.COLOR_RGB2YCrCb)
else:
feature_image = np.copy(img)
# Use cv2.resize().ravel() to create the feature vector
features = cv2.resize(feature_image, size).ravel()
# Return the feature vector
return features
Testing function:
In [494]:
# Read a color image
image = imread('test_images/test4.jpg')
feature_vec = bin_spatial(image, color_space='RGB', size=(32, 32))
# Plot features
plt.plot(feature_vec)
plt.title('Spatially Binned Features')
Out[494]:
HOG Features¶
Extract Histogram of Oriented Gradient features. They are capable of better retaining the uniqueness of the data while also leaving space for variations that account for different representations of the same object:
In [353]:
# Define a function to return HOG features and visualization
def get_hog_features(img, orient, pix_per_cell, cell_per_block,
vis=False, feature_vec=True):
# Call with two outputs if vis==True
if vis == True:
features, hog_image = hog(img, orientations=orient,
pixels_per_cell=(pix_per_cell, pix_per_cell),
cells_per_block=(cell_per_block, cell_per_block),
transform_sqrt=True,
visualise=vis, feature_vector=feature_vec)
return features, hog_image
# Otherwise call with one output
else:
features = hog(img, orientations=orient,
pixels_per_cell=(pix_per_cell, pix_per_cell),
cells_per_block=(cell_per_block, cell_per_block),
transform_sqrt=True,
visualise=vis, feature_vector=feature_vec)
return features
In [495]:
# Read in the image
image = imread('test_images/test4.jpg')
gray = cv2.cvtColor(image, cv2.COLOR_RGB2GRAY)
# Define HOG parameters
orient = 9
pix_per_cell = 8
cell_per_block = 2
# Call our function with vis=True to see an image output
features, hog_image = get_hog_features(gray, orient,
pix_per_cell, cell_per_block,
vis=True, feature_vec=False)
# Plot the examples
fig = plt.figure(figsize=(50, 40))
plt.subplot(121)
plt.imshow(image, cmap='gray')
plt.title('Example Car Image')
plt.subplot(122)
plt.imshow(hog_image, cmap='gray')
plt.title('HOG Visualization')
plt.imsave("output_images/feature_vec.png", hog_image)
Feature Extraction¶
Function to extract features from a list of images, using all of the above histogram methods defined:
In [496]:
def extract_features(imgs, color_space='RGB', spatial_size=(32, 32),
hist_bins=32, orient=9,
pix_per_cell=8, cell_per_block=2, hog_channel=0,
spatial_feat=True, hist_feat=True, hog_feat=True):
# Create a list to append feature vectors to
features = []
# Iterate through the list of images
for file in imgs:
file_features = []
# Read in each one by one
image = imread(file)
# apply color conversion if other than 'RGB'
if color_space != 'RGB':
if color_space == 'HSV':
feature_image = cv2.cvtColor(image, cv2.COLOR_RGB2HSV)
elif color_space == 'LUV':
feature_image = cv2.cvtColor(image, cv2.COLOR_RGB2LUV)
elif color_space == 'HLS':
feature_image = cv2.cvtColor(image, cv2.COLOR_RGB2HLS)
elif color_space == 'YUV':
feature_image = cv2.cvtColor(image, cv2.COLOR_RGB2YUV)
elif color_space == 'YCrCb':
feature_image = cv2.cvtColor(image, cv2.COLOR_RGB2YCrCb)
else: feature_image = np.copy(image)
if spatial_feat == True:
# Apply bin_spatial()
spatial_features = bin_spatial(feature_image, size=spatial_size)
file_features.append(spatial_features)
if hist_feat == True:
# Apply color_hist()
hist_features = color_hist(feature_image, nbins=hist_bins)
file_features.append(hist_features)
if hog_feat == True:
# Call get_hog_features() with vis=False, feature_vec=True
if hog_channel == 'ALL':
hog_features = []
for channel in range(feature_image.shape[2]):
hog_features.append(get_hog_features(feature_image[:,:,channel],
orient, pix_per_cell, cell_per_block,
vis=False, feature_vec=True))
hog_features = np.ravel(hog_features)
else:
hog_features = get_hog_features(feature_image[:,:,hog_channel], orient,
pix_per_cell, cell_per_block, vis=False, feature_vec=True)
# Append the new feature vector to the features list
file_features.append(hog_features)
features.append(np.concatenate(file_features))
# Return list of feature vectors
return features
Sliding Window¶
Define a function that takes an image, start and stop positions in both x and y, window size (x and y dimensions), and overlap fraction (for both x and y), and returns a windows list
In [497]:
def slide_window(img, x_start_stop=[None, None], y_start_stop=[None, None],
xy_window=(64, 64), xy_overlap=(0.5, 0.5)):
# If x and/or y start/stop positions not defined, set to image size
if x_start_stop[0] == None:
x_start_stop[0] = 0
if x_start_stop[1] == None:
x_start_stop[1] = img.shape[1]
if y_start_stop[0] == None:
y_start_stop[0] = 0
if y_start_stop[1] == None:
y_start_stop[1] = img.shape[0]
# Compute the span of the region to be searched
xspan = x_start_stop[1] - x_start_stop[0]
yspan = y_start_stop[1] - y_start_stop[0]
# Compute the number of pixels per step in x/y
nx_pix_per_step = np.int(xy_window[0]*(1 - xy_overlap[0]))
ny_pix_per_step = np.int(xy_window[1]*(1 - xy_overlap[1]))
# Compute the number of windows in x/y
nx_buffer = np.int(xy_window[0]*(xy_overlap[0]))
ny_buffer = np.int(xy_window[1]*(xy_overlap[1]))
nx_windows = np.int((xspan-nx_buffer)/nx_pix_per_step)
ny_windows = np.int((yspan-ny_buffer)/ny_pix_per_step)
# Initialize a list to append window positions to
window_list = []
# Loop through finding x and y window positions
# Note: you could vectorize this step, but in practice
# you'll be considering windows one by one with your
# classifier, so looping makes sense
for ys in range(ny_windows):
for xs in range(nx_windows):
# Calculate window position
startx = xs*nx_pix_per_step + x_start_stop[0]
endx = startx + xy_window[0]
starty = ys*ny_pix_per_step + y_start_stop[0]
endy = starty + xy_window[1]
# Append window position to list
window_list.append(((startx, starty), (endx, endy)))
# Return the list of windows
return window_list
In [499]:
image = imread('test_images/test4.jpg')
windows = slide_window(image, x_start_stop=[None, None], y_start_stop=[None, None],
xy_window=(128, 128), xy_overlap=(0.5, 0.5))
window_img = draw_boxes(image, windows, color=(0, 0, 255), thick=6)
plt.imshow(window_img)
plt.imsave("output_images/sliding_windows.png", window_img)
Vehicle Detection¶
Build the Vehicle Detection Pipeline
Prepare Dataset¶
First we will prepare the dataset for training
Load Image Paths¶
In [377]:
from os import listdir
from os.path import isfile, join
# Data directories names
data_folder = "data"
cars_folder = "vehicles"
notcars_folder = "non-vehicles"
# Load Cars Images
car_paths = "{}/{}/*/*.png".format(data_folder, cars_folder)
cars = glob.glob(car_paths)
print(len(cars))
# Load NonCar Images
notcar_paths = "{}/{}/*/*.png".format(data_folder, notcars_folder)
notcars = glob.glob(notcar_paths)
print(len(notcars))
Dataset characteristic functions:¶
Define a function that computes dataset characteristics:
In [378]:
# Define a function to return some characteristics of the dataset
def data_look(car_list, notcar_list):
data_dict = {}
# Define a key in data_dict "n_cars" and store the number of car images
data_dict["n_cars"] = len(car_list)
# Define a key "n_notcars" and store the number of notcar images
data_dict["n_notcars"] = len(notcar_list)
# Read in a test image, either car or notcar
example_img = imread(car_list[0])
# Define a key "image_shape" and store the test image shape 3-tuple
data_dict["image_shape"] = example_img.shape
# Define a key "data_type" and store the data type of the test image.
data_dict["data_type"] = example_img.dtype
# Return data_dict
return data_dict
Check dataset characteristics:
In [379]:
data_info = data_look(car_list, notcar_list)
print('There are',
data_info["n_cars"], ' cars and',
data_info["n_notcars"], ' non-cars')
print('of size: ',data_info["image_shape"], ' and data type:',
data_info["data_type"])
# Plot one random example for each class
car_ind = np.random.randint(0, len(cars))
notcar_ind = np.random.randint(0, len(notcars))
# Read in car / not-car images
car_image = imread(cars[car_ind])
notcar_image = imread(notcars[notcar_ind])
# Plot the examples
fig = plt.figure()
plt.subplot(121)
plt.imshow(car_image)
plt.title('Example Car Image')
plt.subplot(122)
plt.imshow(notcar_image)
plt.title('Example Not-car Image')
Out[379]:
Extract Features¶
Try out hog feature on an image from each class from the dataset:
In [721]:
# Plot one random example for each class
car_ind = np.random.randint(0, len(cars))
notcar_ind = np.random.randint(0, len(notcars))
# Read in car / not-car images
car_image = imread(cars[car_ind])
notcar_image = imread(notcars[notcar_ind])
# Convert Images to YCrCb Space
ycrcb_car_image = cv2.cvtColor(car_image, cv2.COLOR_RGB2YCrCb)
ycrcb_notcar_image = cv2.cvtColor(notcar_image, cv2.COLOR_RGB2YCrCb)
# Extract Colour Channels
car_ch1 = ycrcb_car_image[:,:,0]
car_ch2 = ycrcb_car_image[:,:,1]
car_ch3 = ycrcb_car_image[:,:,2]
notcar_ch1 = ycrcb_notcar_image[:,:,0]
notcar_ch2 = ycrcb_notcar_image[:,:,1]
notcar_ch3 = ycrcb_notcar_image[:,:,2]
# Define HOG parameters
orient = 11
pix_per_cell = 8
cell_per_block = 2
_, car_ch1_img = get_hog_features(car_ch1, orient,
pix_per_cell, cell_per_block,
vis=True, feature_vec=False)
_, car_ch2_img = get_hog_features(car_ch2, orient,
pix_per_cell, cell_per_block,
vis=True, feature_vec=False)
_, car_ch3_img = get_hog_features(car_ch3, orient,
pix_per_cell, cell_per_block,
vis=True, feature_vec=False)
_, notcar_ch1_img = get_hog_features(notcar_ch1, orient,
pix_per_cell, cell_per_block,
vis=True, feature_vec=False)
_, notcar_ch2_img = get_hog_features(notcar_ch2, orient,
pix_per_cell, cell_per_block,
vis=True, feature_vec=False)
_, notcar_ch3_img = get_hog_features(notcar_ch3, orient,
pix_per_cell, cell_per_block,
vis=True, feature_vec=False)
# Plot Results
f, axs = plt.subplots(4, 4, figsize=(20, 10))
# Original Image with YCrCb Channels
axs[0][0].imshow(car_image)
axs[0][1].imshow(car_ch1)
axs[0][2].imshow(car_ch2)
axs[0][3].imshow(car_ch3)
# YCrCb Image with Channel Gradients
axs[1][0].imshow(ycrcb_car_image)
axs[1][1].imshow(car_ch1_img)
axs[1][2].imshow(car_ch2_img)
axs[1][3].imshow(car_ch2_img)
# Original Image with YCrCb Channels
axs[2][0].imshow(notcar_image)
axs[2][1].imshow(notcar_ch1)
axs[2][2].imshow(notcar_ch2)
axs[2][3].imshow(notcar_ch3)
# YCrCb Image with Channel Gradients
axs[3][0].imshow(ycrcb_notcar_image)
axs[3][1].imshow(notcar_ch1_img)
axs[3][2].imshow(notcar_ch2_img)
axs[3][3].imshow(notcar_ch2_img)
plt.imsave("output_images/car_ch1_img.png", car_ch1_img)
plt.imsave("output_images/car_ch2_img.png", car_ch2_img)
plt.imsave("output_images/car_ch3_img.png", car_ch3_img)
plt.imsave("output_images/notcar_ch1_img.png", notcar_ch1_img)
plt.imsave("output_images/notcar_ch2_img.png", notcar_ch2_img)
plt.imsave("output_images/notcar_ch3_img.png", notcar_ch3_img)
Redefine the extraction functions, modified for the pipeline:
In [380]:
# Define a function to return HOG features and visualization
def get_hog_features(img, orient, pix_per_cell, cell_per_block,
vis=False, feature_vec=True):
# Call with two outputs if vis==True
if vis == True:
features, hog_image = hog(img, orientations=orient,
pixels_per_cell=(pix_per_cell, pix_per_cell),
cells_per_block=(cell_per_block, cell_per_block),
transform_sqrt=True,
visualise=vis, feature_vector=feature_vec)
return features, hog_image
# Otherwise call with one output
else:
features = hog(img, orientations=orient,
pixels_per_cell=(pix_per_cell, pix_per_cell),
cells_per_block=(cell_per_block, cell_per_block),
transform_sqrt=True,
visualise=vis, feature_vector=feature_vec)
return features
# Define a function to compute binned color features
def bin_spatial(img, size=(32, 32)):
# Use cv2.resize().ravel() to create the feature vector
features = cv2.resize(img, size).ravel()
# Return the feature vector
return features
# Define a function to compute color histogram features
# NEED TO CHANGE bins_range if reading .png files with mpimg!
def color_hist(img, nbins=32, bins_range=(0, 256)):
# Compute the histogram of the color channels separately
channel1_hist = np.histogram(img[:,:,0], bins=nbins, range=bins_range)
channel2_hist = np.histogram(img[:,:,1], bins=nbins, range=bins_range)
channel3_hist = np.histogram(img[:,:,2], bins=nbins, range=bins_range)
# Concatenate the histograms into a single feature vector
hist_features = np.concatenate((channel1_hist[0], channel2_hist[0], channel3_hist[0]))
# Return the individual histograms, bin_centers and feature vector
return hist_features
# Define a function to extract features from a list of images
# Have this function call bin_spatial() and color_hist()
def extract_features(imgs, color_space='RGB', spatial_size=(32, 32),
hist_bins=32, orient=9,
pix_per_cell=8, cell_per_block=2, hog_channel=0,
spatial_feat=True, hist_feat=True, hog_feat=True):
# Create a list to append feature vectors to
features = []
# Iterate through the list of images
for file in imgs:
file_features = []
# Read in each one by one
image = imread(file)
# apply color conversion if other than 'RGB'
if color_space != 'RGB':
if color_space == 'HSV':
feature_image = cv2.cvtColor(image, cv2.COLOR_RGB2HSV)
elif color_space == 'LUV':
feature_image = cv2.cvtColor(image, cv2.COLOR_RGB2LUV)
elif color_space == 'HLS':
feature_image = cv2.cvtColor(image, cv2.COLOR_RGB2HLS)
elif color_space == 'YUV':
feature_image = cv2.cvtColor(image, cv2.COLOR_RGB2YUV)
elif color_space == 'YCrCb':
feature_image = cv2.cvtColor(image, cv2.COLOR_RGB2YCrCb)
else: feature_image = np.copy(image)
if spatial_feat == True:
spatial_features = bin_spatial(feature_image, size=spatial_size)
file_features.append(spatial_features)
if hist_feat == True:
# Apply color_hist()
hist_features = color_hist(feature_image, nbins=hist_bins)
file_features.append(hist_features)
if hog_feat == True:
# Call get_hog_features() with vis=False, feature_vec=True
if hog_channel == 'ALL':
hog_features = []
for channel in range(feature_image.shape[2]):
hog_features.append(get_hog_features(feature_image[:,:,channel],
orient, pix_per_cell, cell_per_block,
vis=False, feature_vec=True))
hog_features = np.ravel(hog_features)
else:
hog_features = get_hog_features(feature_image[:,:,hog_channel], orient,
pix_per_cell, cell_per_block, vis=False, feature_vec=True)
# Append the new feature vector to the features list
file_features.append(hog_features)
features.append(np.concatenate(file_features))
# Return list of feature vectors
return features
def convert_color(img, conv='RGB2YCrCb'):
if conv == 'RGB2YCrCb':
return cv2.cvtColor(img, cv2.COLOR_RGB2YCrCb)
if conv == 'BGR2YCrCb':
return cv2.cvtColor(img, cv2.COLOR_BGR2YCrCb)
if conv == 'RGB2LUV':
return cv2.cvtColor(img, cv2.COLOR_RGB2LUV)
HOG Features¶
Extract only hog features for training
In [628]:
### Feacture Extraction Parmeters
hog_feat = True # HOG features on or off
color_space = 'YCrCb' # Can be RGB, HSV, LUV, HLS, YUV, YCrCb
orient = 11 # HOG orientations
hog_channel = 'ALL' # Can be 0, 1, 2, or "ALL"
pix_per_cell = 8 # HOG pixels per cell
cell_per_block = 2 # HOG cells per block
spatial_size = (16, 16) # Spatial binning dimensions
hist_bins = 32 # Number of histogram bins
spatial_feat = False # Spatial features on or off
hist_feat = False # Histogram features on or off
car_features = extract_features(cars, color_space=color_space,
spatial_size=spatial_size, hist_bins=hist_bins,
orient=orient, pix_per_cell=pix_per_cell,
cell_per_block=cell_per_block,
hog_channel=hog_channel, spatial_feat=spatial_feat,
hist_feat=hist_feat, hog_feat=hog_feat)
notcar_features = extract_features(notcars, color_space=color_space,
spatial_size=spatial_size, hist_bins=hist_bins,
orient=orient, pix_per_cell=pix_per_cell,
cell_per_block=cell_per_block,
hog_channel=hog_channel, spatial_feat=spatial_feat,
hist_feat=hist_feat, hog_feat=hog_feat)
Normalize Data¶
In [629]:
X = np.vstack((car_features, notcar_features)).astype(np.float64)
# Fit a per-column scaler
X_scaler = StandardScaler().fit(X)
# Apply the scaler to X
scaled_X = X_scaler.transform(X)
Define Labels¶
In [630]:
# Define the labels vector
y = np.hstack((np.ones(len(car_features)), np.zeros(len(notcar_features))))
Split data into training and testing¶
In [709]:
# Split up data into randomized training and test sets
rand_state = np.random.randint(0, 100)
X_train, X_test, y_train, y_test = train_test_split(
scaled_X, y, test_size=0.2, random_state=rand_state)
print('Using:',orient,'orientations',pix_per_cell,
'pixels per cell and', cell_per_block,'cells per block')
print('Feature vector length:', len(X_train[0]))
Traing Model¶
Train the model that we will use for detections
In [632]:
# Use a linear SVC
svc = LinearSVC()
# Check the training time for the SVC
t=time.time()
svc.fit(X_train, y_train)
t2 = time.time()
print(round(t2-t, 2), 'Seconds to train SVC...')
Test Accuracy
In [633]:
# Check the score of the SVC
print('Test Accuracy of SVC = ', round(svc.score(X_test, y_test), 4))
Using Grid Search for Parameters Tunning¶
Tune the hyperparameters using grid search:
In [160]:
from sklearn.grid_search import GridSearchCV
parameters = {'loss':('hinge', 'squared_hinge'), 'tol': [1e-1, 1e-2, 1e-3, 1e-4, 1e-5, 1e-6], 'C':[0.5, 1, 1.5, 2]}
svm_linear = LinearSVC()
clf = GridSearchCV(svm_linear, parameters)
clf.fit(X_train, y_train)
Out[160]:
Check best params:
Print best params:
In [164]:
clf.best_params_
Out[164]:
Check accuracy with this configuration:
In [167]:
# Use a linear SVC
optim_svc = LinearSVC(C=1, loss='squared_hinge', tol=0.1)
# Check the training time for the SVC
t=time.time()
optim_svc.fit(X_train, y_train)
t2 = time.time()
print(round(t2-t, 2), 'Seconds to train SVC...')
# Check the score of the SVC
print('Test Accuracy of SVC = ', round(optim_svc.score(X_test, y_test), 4))
Sliding Window Search¶
Define a function that takes an image, start and stop positions in both x and y, window size (x and y dimensions), and overlap fraction (for both x and y), and returns a list of windows:
In [634]:
def slide_window(img, x_start_stop=[None, None], y_start_stop=[None, None],
xy_window=(64, 64), xy_overlap=(0.5, 0.5)):
# If x and/or y start/stop positions not defined, set to image size
if x_start_stop[0] == None:
x_start_stop[0] = 0
if x_start_stop[1] == None:
x_start_stop[1] = img.shape[1]
if y_start_stop[0] == None:
y_start_stop[0] = 0
if y_start_stop[1] == None:
y_start_stop[1] = img.shape[0]
# Compute the span of the region to be searched
xspan = x_start_stop[1] - x_start_stop[0]
yspan = y_start_stop[1] - y_start_stop[0]
# Compute the number of pixels per step in x/y
nx_pix_per_step = np.int(xy_window[0]*(1 - xy_overlap[0]))
ny_pix_per_step = np.int(xy_window[1]*(1 - xy_overlap[1]))
# Compute the number of windows in x/y
nx_buffer = np.int(xy_window[0]*(xy_overlap[0]))
ny_buffer = np.int(xy_window[1]*(xy_overlap[1]))
nx_windows = np.int((xspan-nx_buffer)/nx_pix_per_step)
ny_windows = np.int((yspan-ny_buffer)/ny_pix_per_step)
# Initialize a list to append window positions to
window_list = []
# Loop through finding x and y window positions
# Note: you could vectorize this step, but in practice
# you'll be considering windows one by one with your
# classifier, so looping makes sense
for ys in range(ny_windows):
for xs in range(nx_windows):
# Calculate window position
startx = xs*nx_pix_per_step + x_start_stop[0]
endx = startx + xy_window[0]
starty = ys*ny_pix_per_step + y_start_stop[0]
endy = starty + xy_window[1]
# Append window position to list
window_list.append(((startx, starty), (endx, endy)))
# Return the list of windows
return window_list
Define a function to extract features from a single image. This function is very similar to extract_features(), but it just extracts features for a single image rather than a list of images.
In [635]:
def single_img_features(img, color_space='RGB', spatial_size=(32, 32),
hist_bins=32, orient=9,
pix_per_cell=8, cell_per_block=2, hog_channel=0,
spatial_feat=True, hist_feat=True, hog_feat=True):
# 1) Define an empty list to receive features
img_features = []
# 2) Apply color conversion if other than 'RGB'
if color_space != 'RGB':
if color_space == 'HSV':
feature_image = cv2.cvtColor(img, cv2.COLOR_RGB2HSV)
elif color_space == 'LUV':
feature_image = cv2.cvtColor(img, cv2.COLOR_RGB2LUV)
elif color_space == 'HLS':
feature_image = cv2.cvtColor(img, cv2.COLOR_RGB2HLS)
elif color_space == 'YUV':
feature_image = cv2.cvtColor(img, cv2.COLOR_RGB2YUV)
elif color_space == 'YCrCb':
feature_image = cv2.cvtColor(img, cv2.COLOR_RGB2YCrCb)
else:
feature_image = np.copy(img)
# 3) Compute spatial features if flag is set
if spatial_feat == True:
spatial_features = bin_spatial(feature_image, size=spatial_size)
# 4) Append features to list
img_features.append(spatial_features)
# 5) Compute histogram features if flag is set
if hist_feat == True:
hist_features = color_hist(feature_image, nbins=hist_bins)
#6) Append features to list
img_features.append(hist_features)
# 7) Compute HOG features if flag is set
if hog_feat == True:
if hog_channel == 'ALL':
hog_features = []
for channel in range(feature_image.shape[2]):
hog_features.extend(get_hog_features(feature_image[:,:,channel],
orient, pix_per_cell, cell_per_block,
vis=False, feature_vec=True))
else:
hog_features = get_hog_features(feature_image[:,:,hog_channel], orient,
pix_per_cell, cell_per_block, vis=False, feature_vec=True)
# 8) Append features to list
img_features.append(hog_features)
# 9) Return concatenated array of features
return np.concatenate(img_features)
Define a function you will pass an image and the list of windows to be searched (output of slide_windows())
In [636]:
def search_windows(img, windows, clf, scaler, color_space='RGB',
spatial_size=(32, 32), hist_bins=32,
hist_range=(0, 256), orient=9,
pix_per_cell=8, cell_per_block=2,
hog_channel=0, spatial_feat=True,
hist_feat=True, hog_feat=True):
# 1) Create an empty list to receive positive detection windows
on_windows = []
# 2) Iterate over all windows in the list
for window in windows:
# 3) Extract the test window from original image
test_img = cv2.resize(img[window[0][1]:window[1][1], window[0][0]:window[1][0]], (64, 64))
# 4) Extract features for that window using single_img_features()
features = single_img_features(test_img, color_space=color_space,
spatial_size=spatial_size, hist_bins=hist_bins,
orient=orient, pix_per_cell=pix_per_cell,
cell_per_block=cell_per_block,
hog_channel=hog_channel, spatial_feat=spatial_feat,
hist_feat=hist_feat, hog_feat=hog_feat)
# 5) Scale extracted features to be fed to classifier
test_features = scaler.transform(np.array(features).reshape(1, -1))
#6) Predict using your classifier
prediction = clf.predict(test_features)
#7) If positive (prediction == 1) then save the window
if prediction == 1:
on_windows.append(window)
#8) Return windows for positive detections
return on_windows
Test Classifier on Image¶
In [723]:
# Check the prediction time for a single sample
t=time.time()
image = imread('test_images/test1.jpg')
draw_image = np.copy(image)
small_windows = slide_window(image, x_start_stop=[None, None], y_start_stop=[350, 500],
xy_window=(64, 64), xy_overlap=(0.7, 0.7))
medium_windows = slide_window(image, x_start_stop=[None, None], y_start_stop=[450, 600],
xy_window=(128, 128), xy_overlap=(0.7, 0.7))
large_windows = slide_window(image, x_start_stop=[None, None], y_start_stop=[500, None],
xy_window=(256, 256), xy_overlap=(0.7, 0.7))
windows = small_windows + medium_windows + large_windows
hot_windows = search_windows(image, windows, svc, X_scaler, color_space=color_space,
spatial_size=spatial_size, hist_bins=hist_bins,
orient=orient, pix_per_cell=pix_per_cell,
cell_per_block=cell_per_block,
hog_channel=hog_channel, spatial_feat=spatial_feat,
hist_feat=hist_feat, hog_feat=hog_feat)
window_img = draw_boxes(draw_image, hot_windows, color=(0, 0, 255), thick=6)
plt.imshow(window_img)
plt.imsave("output_images/testing_classifier_1.png", window_img)
HOG Sub-sampling Window search¶
Function that can extract features using hog sub-sampling and makes predictions
In [740]:
import numpy as np
import cv2
from skimage.feature import hog
def convert_color(img, color_space='YCrCb'):
feature_image = np.copy(img)
# 2) Apply color conversion if other than 'RGB'
if color_space != 'RGB':
if color_space == 'HSV':
feature_image = cv2.cvtColor(img, cv2.COLOR_RGB2HSV)
elif color_space == 'LUV':
feature_image = cv2.cvtColor(img, cv2.COLOR_RGB2LUV)
elif color_space == 'HLS':
feature_image = cv2.cvtColor(img, cv2.COLOR_RGB2HLS)
elif color_space == 'YUV':
feature_image = cv2.cvtColor(img, cv2.COLOR_RGB2YUV)
elif color_space == 'YCrCb':
feature_image = cv2.cvtColor(img, cv2.COLOR_RGB2YCrCb)
return feature_image
def find_cars(img, ystart, ystop, scale, svc, X_scaler, orient, pix_per_cell, cell_per_block, draw_on_image=True):
if draw_on_image:
draw_img = np.copy(img)
img_tosearch = img[ystart:ystop, :, :]
# Apply Conversion
ctrans_tosearch = convert_color(img_tosearch, color_space='YCrCb')
# Rescale Image
if scale != 1:
imshape = ctrans_tosearch.shape
ctrans_tosearch = cv2.resize(ctrans_tosearch, (np.int(imshape[1]/scale), np.int(imshape[0]/scale)))
# Extract Colour Channels
ch1 = ctrans_tosearch[:,:,0]
ch2 = ctrans_tosearch[:,:,1]
ch3 = ctrans_tosearch[:,:,2]
# Define blocks and steps as above
nxblocks = (ch1.shape[1] // pix_per_cell) - cell_per_block + 1
nyblocks = (ch1.shape[0] // pix_per_cell) - cell_per_block + 1
nfeat_per_block = orient * cell_per_block**2
# 64 was the orginal sampling rate, with 8 cells and 8 pix per cell
window = 64
nblocks_per_window = (window // pix_per_cell) - cell_per_block + 1
cells_per_step = 2 # Instead of overlap, define how many cells to step
nxsteps = (nxblocks - nblocks_per_window) // cells_per_step + 1
nysteps = (nyblocks - nblocks_per_window) // cells_per_step + 1
# Compute individual channel HOG features for the entire image
hog1 = get_hog_features(ch1, orient, pix_per_cell, cell_per_block, feature_vec=False)
hog2 = get_hog_features(ch2, orient, pix_per_cell, cell_per_block, feature_vec=False)
hog3 = get_hog_features(ch3, orient, pix_per_cell, cell_per_block, feature_vec=False)
# Array to store boxes
bboxes = []
for xb in range(nxsteps):
for yb in range(nysteps):
ypos = yb * cells_per_step
xpos = xb * cells_per_step
# Extract HOG for this patch
hog_feat1 = hog1[ypos:ypos+nblocks_per_window, xpos:xpos+nblocks_per_window].ravel()
hog_feat2 = hog2[ypos:ypos+nblocks_per_window, xpos:xpos+nblocks_per_window].ravel()
hog_feat3 = hog3[ypos:ypos+nblocks_per_window, xpos:xpos+nblocks_per_window].ravel()
hog_features = np.hstack((hog_feat1, hog_feat2, hog_feat3))
xleft = xpos*pix_per_cell
ytop = ypos*pix_per_cell
# Scale features and make a prediction
test_features = X_scaler.transform(hog_features.reshape(1, -1))
test_prediction = svc.predict(test_features)
if test_prediction == 1:
xbox_left = np.int(xleft*scale)
ytop_draw = np.int(ytop*scale)
win_draw = np.int(window*scale)
# Append Box
bboxes.append(((xbox_left, ytop_draw+ystart),(xbox_left+win_draw,ytop_draw+win_draw+ystart)))
# Flag for speeding processing on video
if draw_on_image:
cv2.rectangle(draw_img,(xbox_left, ytop_draw+ystart),(xbox_left+win_draw,ytop_draw+win_draw+ystart), (0,0,255),6)
if draw_on_image:
return bboxes, draw_img
else:
return bboxes
Test It
In [741]:
# Get Test Images Path
test_images = glob.glob('test_images/*.jpg')
# Tune Params
ystart = 350
ystop = 656
scale = 1.2
cell_per_block = 2
pix_per_cell = 8
orient = 11
f, axs = plt.subplots(3, 2, figsize=(20, 10))
axs = axs.ravel()
i = 0
# Process all Images
for img in test_images:
# Read Image
tst_img = imread(img)
# Find Cars
boxes, out_img = find_cars(tst_img, ystart, ystop, scale, svc, X_scaler, orient, pix_per_cell, cell_per_block)
# Display Boxes
axs[i].imshow(out_img)
# Get Image Name
image_name = os.path.basename(img)
# Save Image
plt.imsave("output_images/{}-predicted.png".format(image_name[:-4]), out_img)
i += 1
plt.show()
Multiple Detections & False Positives¶
Create a heat-map from detections in order to combine overlapping detections and remove false positives.
Add heat (+1) for all pixesl within boxes:
In [742]:
def add_heat(heatmap, bbox_list):
# Iterate through list of bboxes
for box in bbox_list:
# Add += 1 for all pixels inside each bbox
# Assuming each "box" takes the form ((x1, y1), (x2, y2))
heatmap[box[0][1]:box[1][1], box[0][0]:box[1][0]] += 1
# Return updated heatmap
return heatmap
Imposing a threshold to reject areas affected by false positives:
In [743]:
def apply_threshold(heatmap, threshold):
# Zero out pixels below the threshold
heatmap[heatmap <= threshold] = 0
# Return thresholded map
return heatmap
Function to draw bounding boxes around the labeled regions:
In [744]:
def draw_labeled_bboxes(img, labels):
# Iterate through all detected cars
for car_number in range(1, labels[1]+1):
# Find pixels with each car_number label value
nonzero = (labels[0] == car_number).nonzero()
# Identify x and y values of those pixels
nonzeroy = np.array(nonzero[0])
nonzerox = np.array(nonzero[1])
# Define a bounding box based on min/max x and y
bbox = ((np.min(nonzerox), np.min(nonzeroy)), (np.max(nonzerox), np.max(nonzeroy)))
# Draw the box on the image
cv2.rectangle(img, bbox[0], bbox[1], (0,0,255), 6)
# Return the image
return img
Test It on above Image
In [760]:
f, axs = plt.subplots(6, 3, figsize=(40, 30))
axs = axs.ravel()
i = 0
# Process all Images
for img in test_images:
# Read Image
tst_img = imread(img)
# Find Cars
box_list, out_img = find_cars(tst_img, ystart, ystop, scale, svc, X_scaler, orient, pix_per_cell, cell_per_block)
# Prepare heat map
heat = np.zeros_like(out_img[:,:,0]).astype(np.float)
# Add heat to each box in box list
heat = add_heat(heat, box_list)
# Apply threshold to help remove false positives
heat = apply_threshold(heat, 1)
# Visualize the heatmap when displaying
heatmap = np.clip(heat, 0, 255)
# Find final boxes from heatmap using label function
labels = label(heatmap)
draw_img = draw_labeled_bboxes(np.copy(tst_img), labels)
# Display Boxes
axs[i].imshow(draw_img)
# Display Labels
axs[i + 1].imshow(labels[0])
# Display Heatmap
axs[i + 2].imshow(heatmap)
i += 3
plt.show()
Put them all together in a single noise filter function:
In [761]:
def filter_noise(box_lists, heat, thresh=1):
boxes = []
for box_list in box_lists:
boxes = boxes + box_list
# Add heat to each box in box list
heat = add_heat(heat, boxes)
# Apply threshold to help remove false positives
heat = apply_threshold(heat, thresh)
# Visualize the heatmap when displaying
heatmap = np.clip(heat, 0, 255)
# Find final boxes from heatmap using label function
labels = label(heatmap)
return labels, heatmap
Integrate a heat map over several frames of video, such that areas of multiple detections get "hot", while transient false positives stay "cool".
Vehicle Detection Pipeline:¶
Now we can put everything together in a single vehicle detection pipeline:
In [775]:
# Tune Params
ystart = 350
ystop = 656
scale = 1.2
cell_per_block = 2
pix_per_cell = 8
orient = 11
class Counter():
def __init__(self, n, thresh):
self.MAX = n
self.box_list = []
self.thresh = thresh
def pipeline(img):
image = np.copy(img)
current_boxes = find_cars(image, ystart, ystop, scale, svc, X_scaler, orient, pix_per_cell, cell_per_block, draw_on_image=False)
if len(counter.box_list) > counter.MAX:
# FIFO stack of boxes
counter.box_list = counter.box_list[1:]
counter.box_list.append(current_boxes)
# Prepare heat map
heat = np.zeros_like(image[:,:,0]).astype(np.float)
labels, heatmap = filter_noise(counter.box_list, heat, thresh=counter.thresh)
# Find final boxes from heatmap using label function
labels = label(heatmap)
draw_img = draw_labeled_bboxes(np.copy(image), labels)
return draw_img
Test on Videos¶
In [763]:
# Import everything needed to edit/save/watch video clips
from moviepy.editor import VideoFileClip
from IPython.display import HTML
Test Video¶
In [792]:
counter = Counter(2, 1)
test_output = 'output_videos/test_video_output.mp4'
test_clip = VideoFileClip("test_video.mp4")
processed_test_clip = test_clip.fl_image(pipeline) #NOTE: this function expects color images!!
%time processed_test_clip.write_videofile(test_output, audio=False)
In [794]:
HTML("""
<video width="960" height="540" controls>
<source src="{0}">
</video>
""".format(test_output))
Out[794]:
Project Video¶
In [767]:
counter = Counter(2, 1)
project_output = 'output_videos/project_video_output.mp4'
project_clip = VideoFileClip("project_video.mp4")
processed_project_clip = project_clip.fl_image(pipeline) #NOTE: this function expects color images!!
%time processed_project_clip.write_videofile(project_output, audio=False)
In [768]:
HTML("""
<video width="960" height="540" controls>
<source src="{0}">
</video>
""".format(project_output))
Out[768]:
In [795]:
counter = Counter(4, 3)
project_output_2 = 'output_videos/project_video_output_2.mp4'
##clip1 = VideoFileClip("test_videos/solidWhiteRight.mp4").subclip(0,5)
project_clip_2 = VideoFileClip("project_video.mp4").subclip(0,10)
processed_project_clip_2 = project_clip_2.fl_image(pipeline) #NOTE: this function expects color images!!
%time processed_project_clip_2.write_videofile(project_output_2, audio=False)
In [796]:
HTML("""
<video width="960" height="540" controls>
<source src="{0}">
</video>
""".format(project_output_2))
Out[796]:
My Video¶
In [790]:
counter = Counter(4, 2)
my_video = 'output_videos/my_video_output.mp4'
##clip1 = VideoFileClip("test_videos/solidWhiteRight.mp4").subclip(0,5)
my_clip = VideoFileClip("my_video.mov")
processed_my_clip = my_clip.fl_image(pipeline) #NOTE: this function expects color images!!
%time processed_my_clip.write_videofile(my_video, audio=False)
In [791]:
HTML("""
<video width="960" height="540" controls>
<source src="{0}">
</video>
""".format(my_video))
Out[791]: